Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: Elmer who wrote (73430)	9/29/1999 12:28:00 PM
From: kash johal	Read Replies (2) \| Respond to of 1573428

Elmer, Re:"It would seem far simplier to me to guardband the VCC rather than the frequency, and it would accomplish the same thing. In an attempt to keep test time at a minimum, the question would be, which requires the least time? A voltage switch or a reload of the timing registers? With the constant concern of violating tester timing restrictions, having multiple timing templates could be a major resource consuming task. Furthermore, going to 800Meg or 1Gig core speed would require ultra high speeds on the cache bus even at only 1/2 core speeds. Extremely difficult to test in a production environment. All this points out the enormous value of low defect density, allowing large on-die caches. No need to test the BSB because there isn't any. Much lower tester speeds and much lower pin count thus much lower capitol costs. " Boy you shure are desperate with the fud. Seems to me your BSB needs some testing too. The Athlon die are NOT tested at speed on wafer form. The speed testing is done with the cache on the module. At final test the module is then rated at the appropriate speed. As far as guardbanding VCC etc. The devices are tested to ensure they meet the specs for each speed grade. So for a 2.0V device the spec for min amd max may be 1.9V to 2.1V. The tester will test to 1.8V to 2.15V there is NO need for 5-10% margins just what is adequate for the tester capability. This whole guardbanding discussion of 5-10% is silly frankly. The market and PLLs and MB determine the appropriate clock speeds the devices need to be tested at. If the spec is 500Mhz, they may test at 505Mhz as min to ensure tester accuracy doesnot come into product binning, but no way are 500Mhz parts really 550Mhz parts. Now clearly there is a speed distribution and within the 500Mhz speed grade the actual speed limit of the part will fall within 500Mhz and the next upper speed grades test limit. And finally the test time on the CPU market is largely irrelevant. The cost of 100UPH or 2500UPH is basically a dollar or two. regards, Kash